基于DBSCAN算法的海战场环境下目标分群  

Target clustering in naval battlefield environment based on DBSCAN algorithm

作　　者：吴艳杰 薛涵 尹东亮 WU Yanjie;XUE Han;YIN Dongliang(Dept.of Operational Research and Programming,Naval Univ.of Engineering,Wuhan 430033,China)

机构地区：[1]海军工程大学作战运筹与规划系,武汉430033

出　　处：《海军工程大学学报》2023年第4期71-75,共5页Journal of Naval University of Engineering

基　　金：海军工程大学自主立项基金资助项目(2022509010)。

摘　　要：针对海战场环境下态势评估中目标数量多、类型复杂多样的问题,首先引入数据聚类对态势评估的目标分群环节进行聚类分群,提出基于DBSCAN(density-based spatial clustering of applications with noise)算法的密度聚类,可聚类任意形状的数据簇,遍历性好,能够对战场环境下目标进行全面合理的分群;然后,给出了算法计算的基本步骤,并利用算例对已知战场态势的目标群进行正确性验证;最后,将该算法与基于划分的K-means算法、基于层次的AGNES(AGglomerative NESting)算法进行了对比分析,证明了该算法的有效性和合理性。Aiming at the problems of large number and complex types of targets in the situation assessment in the naval battlefield environment,data clustering was introduced to cluster target cluster link of the situation assessment.Density-based spatial clustering of applications with noise algorithm was proposed,by which data clusters of arbitrary shapes can be clustered with good ergodicity,and meanwhile targets under battlefield environment can be clustered in a comprehensive and reasonable way.The basic steps of algorithm calculation were given.And numerical examples were used to verify the correctness of target groups with known battlefield situations.In addition,the new algorithm was compared with K-means algorithm based on partition and AGglomerative NESting algorithm,and thus its validity and rationality were proved.

关 键 词：数据聚类 DBSCAN算法 密度 目标分群 海战场环境 

分 类 号：F253.4[经济管理—国民经济]